Dawn Journal: How does Dawn know where "down" is?

Dear Asdawnished Readers,

Dawn is scrutinizing Vesta from its low-altitude mapping orbit (LAMO), circling the rocky world five and a half times a day. The spacecraft is healthy and continuing its intensive campaign to reveal the astonishing nature of this body in the mysterious depths of the main asteroid belt. Since the last log, the robotic explorer has devoted most of its time to its two primary scientific objectives in this phase of the mission. With its gamma ray and neutron detector (GRaND), it has been patiently measuring Vesta's very faint nuclear emanations. These signals reveal the atomic constituents of the material near the surface. Dawn also broadcasts a radio beacon with which navigators on distant Earth can track its orbital motion with exquisite accuracy. That allows them to measure Vesta's gravity field and thereby infer the interior structure of this complex world. In addition to these top priorities, the spacecraft is using its camera and its visible and infrared mapping spectrometer (VIR) to obtain more detailed views than they could in the higher orbits.

As we have delved into these activities in detail in past logs, let's consider here some more aspects of controlling this extremely remote probe as it peers down at the exotic colossus 210 kilometers (130 miles) beneath it.

Well, the first aspect that is worth noting is that it is incredibly cool! Continuing to bring this fascinating extraterrestrial orb into sharper focus is thrilling, and everyone who is moved by humankind's bold efforts to reach into the cosmos shares in the experience. As a reminder, you can see the extraordinary sights Dawn has by going here for a new image every weekday, each revealing another intriguing aspect of the diverse landscape.

The data sent back are providing exciting and important new insights into Vesta, and those findings will continue to be announced in press releases. Therefore, we will turn our attention to a second aspect of operating in LAMO. Last month, we saw that various forces contribute to Dawn moving slightly off its planned orbital path. (That material may be worth reviewing, either to enhance appreciation of what follows or as an efficacious soporific, should the need for one ever arise.) Now let's investigate some of the consequences. This will involve a few more technical points than most logs, but each will be explained, and together they will help illustrate one of the multitudinous complexities that must be overcome to make such a grand adventure successful.

Far away, traveling through the vast expanse of (mostly) empty space, Dawn only knows where it is because of information the mission control team installs in it. This is typical for interplanetary spacecraft. Earth-orbiting satellites may be able to use the Global Positioning System (GPS) constellation or other means to find their own location, but only a few spacecraft that have gone far from Earth have the means to independently establish their own location. This should not be confused with a spacecraft's ability to determine its own orientation, which Dawn does with its star trackers, gyros, and sun sensors. In the same way, if you were in a dark and unidentified place on your planet, you could determine the direction you were looking by recognizing patterns of stars, but that would not help you ascertain your position.

Throughout the mission, controllers regularly transmit to the spacecraft a mathematical description of its location in the solar system at any instant over a given period of time. They also provide it with the information needed to calculate where Earth is. That's how it is able to point its main antenna in the correct direction when it needs to do so. During the Vesta phase of the mission, the probe is given the additional means it needs to determine its location relative to Vesta. All the information sent to the spacecraft is based on navigators' best prediction of where the spacecraft will be in the future. Dawn remains unaware of any deviations from its expected course, so it always behaves as if it were exactly where it would be if its motion matched the team's projections perfectly, without the discrepancies that are sure to occur. For the majority of the mission, both in interplanetary cruise and at higher altitude orbits at Vesta, the effects of being slightly off the predicted trajectory are insignificant. In LAMO, they are not.

For Dawn to aim its scientific sensors at Vesta, controllers instruct it to point straight "down." Again, it knows how to compute where "down" is because of the information it was given by navigators. Any disparity between where the craft was predicted to be and where it really is along its orbit causes it to point in a slightly different direction, not quite truly straight down. This does not compromise the observations; it could tolerate larger pointing errors and still capture the desired targets in the field of view of the instruments.

Dawn is a very large spacecraft. Indeed, the wingspan from one solar array tip to the other is 19.7 meters (nearly 65 feet). When it was launched in 2007, this was the greatest span of any probe NASA had ever dispatched on an interplanetary journey. The large area of solar cells is needed to capture faint sunlight in the asteroid belt to meet all of the electrical power needs. Each solar array wing is the width of a singles' tennis court, and the whole spacecraft would reach from a pitcher's mound to home plate on a professional baseball field, although Dawn is engaged in activities considerably more inspiring and rewarding than competitive sports.

Now consider that when Dawn is looking precisely down, directly toward the center of Vesta, its wings are level. If it is pointed off even a little, then one of those long extensions is slightly closer to the massive body it is circling and one is slightly farther away. Because gravity diminishes with increasing distance, the one that is closer is subject to a very slightly stronger pull than the farther other. If unchecked, that lower side would gently be pulled down even more, thus increasing the difference in gravitational attraction between the two wings still more. Eventually, this would cause Dawn to be oriented so that one wing points straight down toward the ancient surface below and the other points straight up, back into the depths of space. Because this phenomenon depends on the change in gravity from the lower point to the higher one, it is known as "gravity gradient." Some satellites that orbit Earth are designed to take advantage of the gravity gradient to align their long axis with the planet below, but Dawn (and most other spacecraft) need greater flexibility in where they point.

Rather than accepting the passive method of orienting provided by the gravity gradient, Dawn uses its reaction wheels to train its science instruments on Vesta. By electrically changing the rate at which these devices spin, the ship can control its orientation in the zero-gravity, frictionless conditions of spaceflight. When a small deviation from the perfect orbit causes it to tip its wings a little when pointing to where it calculates "down" to be, the spacecraft's reaction wheels work to prevent it from succumbing to the gravity gradient, countering the tendency of the wings to deviate still more from being level. As a consequence, the ship remains stable and the wheels gradually spin faster and faster as it conducts its observations.

To reduce the wheels' speeds, mission planners schedule a period almost every day in LAMO during which the spacecraft fires its reaction control system thrusters, a function known as "desaturating the wheels." Indeed, the principal reason Dawn is outfitted with these small thrusters and a modest supply of conventional rocket propellant known as hydrazine is to manage the speed of the wheels.

The thruster firings not only provide the torque needed to reduce the rotation rate of the wheels, but they also have the incidental effect of propelling the spacecraft slightly. The push is small, changing the orbital speed by no more than about one centimeter per second (around one fiftieth of a mph, or about 120 feet per hour). But that causes Dawn to deviate from its planned orbit, and the accumulated force from all the firings is the largest source of trajectory discrepancies in LAMO.

To summarize so far, once Dawn has any variance at all between the predicted orbital motion that mission controllers have radioed to it and its actual path, its long wings will be tipped a little while it observes Vesta. In opposing the resultant gravity gradient effect, the reaction wheels will accelerate. When the reaction control system thrusters fire to decelerate the wheels, they will nudge Dawn still more off course, and the cycle will continue.

Of course, engineers have devised strategies to accommodate this contribution (and others) to deviations from the plan. In LAMO, they frequently measure the ship's trajectory and revise their estimates of the future course. They transmit to the spacecraft a new prediction for the orbit twice a week, so the main computer usually has a very good estimate of where it is relative to Vesta and hence how to orient itself so that its long solar arrays remain level as it acquires its fabulous pictures and other scientific information. With the updated knowledge of its position, Dawn can aim its sensors accurately and keep the thruster firings from being excessive, even when it is not following its orbit perfectly. This solution works well, but let's continue delving into the consequences of the orbital perturbations.

While the operations team has the capability to provide the ship regularly with a good description of where it will be, it is much more difficult to make such frequent adjustments to its detailed itinerary. The schedule of its myriad activities has to be planned longer in advance. The sequences of commands, which are timed to the second, are very complicated to develop and verify, and the operations team does not have the resources to refine the timing as often as they can send updates on the craft's predicted location.

Engineers took many factors into account in selecting the orbits Dawn uses for its science observations. We saw in November that the orbits are characterized not only by the altitude but also by the orientation of the orbital plane. A subsequent log will explain the choices for the planes more fully, but for now, what matters is that, among other considerations, the orbits were designed to ensure Dawn remains in constant sunlight. It always has the Sun in sight, never entering Vesta's shadow. Keeping Earth in view at all times was not part of the design, and on every one of the more than 600 revolutions around the gigantic rocky body since August 28 (the seventh circuit in survey orbit), the spacecraft has been temporarily behind Vesta from the geocentric point of view. In its present orbit, these occultations last for about half an hour in every 4.3-hour loop.

When Dawn is observing Vesta, that doesn't matter. When it is using its ion propulsion system to transfer from one orbit to another, it also doesn't matter. It does matter, however, when it is in contact with Earth, because Vesta blocks the radio signal. Controllers give the spacecraft a detailed schedule of which data to transmit and when, making the best possible use of the precious communications link that stretches across the solar system. The timed plan tells it not to send high priority data during the radio blackout, but the timing of the occultations can shift a little as the orbit departs from the plan.

The strategy to deal with the slight deviations in the timing of the interruption in the radio link principally involves including some padding in the plan. The schedule for the transmission of the highest priority data places it well away from the expected gap, so no important losses occur if Dawn is a little ahead in its orbit or a little behind (causing the gap to occur a little earlier or a little later).

But what is there to do during and near the time the craft is predicted to be blocked by Vesta while conducting a communications session? Dawn rotates too slowly to make it worth turning to point its sensors at the surface just for these periods. Of course, it could simply transmit nothing at all. Instead, the team has it transmit data that otherwise would be lost. There is never enough time to send to Earth all the information the probe generates and collects. So most of the time it is behind Vesta, it broadcasts many of the measurements of its own subsystems that cannot be stored and sent later. And during the periods immediately before and after the expected occultation, when there is a chance that the signal will reach Earth, it sends bonus pictures and VIR spectra. If the deviations from the planned orbit are small, then the antenna will have an unobstructed view of Earth, and these data will make it home. And if the spacecraft enters the blackout period late (or early), then it will exit late (or early) as well, so the bonus results sent before (or after) the occultation will be received. But in the rest of the cases, well, Dawn will transmit those bits right back where they came from, sending the photos and spectra into the vast rocky surface between the spacecraft and Earth. Last month we described one of the limitations in how much bonus information could be obtained from LAMO. Now we have another. In summary, because the probe can acquire more images and other data than it is possible to return, it radios some of them during times that it is possible they will make it to Earth. Because of realistic causes of variation from its predicted orbital path, however, some of these measurements will be transmitted when, from Dawn's perspective, Vesta blocks Earth, thus preventing the broadcast signals from getting through. The GRaND observations (as well as essential telemetry on the health of the ship) are scheduled to be sent during times that, even with the reasonable range of orbit discrepancies, the communications link will not be obstructed. In this way, mission planners return as much data as possible, taking maximum advantage of the time Dawn points its main antenna to Earth. Having a sophisticated robot in orbit around the second most massive resident of the asteroid belt presents truly unique opportunities for the exploration of the solar system, and the team has devised every strategy they could to use the time as productively as possible.

The spacecraft aims GRaND at Vesta most of the time in order to develop a good picture of the weak nuclear glow. Controllers schedule three periods per week, each about eight hours, in which it directs its antenna to Earth. The orbit predictions have been extremely good, matching the actual motion quite well. Moreover, some time is allocated to return the camera and VIR data apart from the times that Vesta might be in the way. As a result, the team has been rewarded with more than 3200 photos from LAMO so far. Every one is bonus, and every one is neat!

After well over four years of travel in deep space and already half a year in orbit around Vesta, engineers recently encountered a bug lurking in the spacecraft's software. As with most bugs, this one had waited silently until just the right circumstances occurred to provoke it. The combination of conditions was achieved late in the day on January 13, and the bug caused the main computer to reboot. Dawn correctly responded by going into safe mode. The mission control team observed this the next day, and promptly began investigating the reason. They soon determined the nature of the bug (as well as ways to ensure it would never be activated again) and restored the spacecraft to its usual operating configuration for LAMO. Even with the slow communications in safe mode, the long time for radio signals to travel between Earth and Dawn, and the frequent interruptions by the regular occultations by Vesta, they had fully restored all systems by January 19. It took a few more days to configure GRaND, but it, along with the other instruments, is now back to its intensive inspection of Vesta.

We saw last month that the mission has been progressing so well that the time originally allocated to deal with anomalies had not been needed, so it is being applied to extend the duration of LAMO. This allows even more scientific observations to be conducted in this lowest altitude. Far from the planet it left in 2007, in a region of the solar system in which no other spacecraft has ever taken up residence, Dawn will continue its exploration of Vesta, alternating between examining the alien world below and transmitting its discoveries to Earth. Meanwhile, everyone who ponders what undiscovered lands lie beyond our sight, everyone who hungers for exciting challenges and noble adventures, and everyone who values turning the unknown into the known profits from the great treasures this stalwart cosmic ambassador sends to its erstwhile home, a faraway place it will never visit again.

Dawn is 210 kilometers (130 miles) from Vesta. It is also 3.08 AU (461 million kilometers or 286 million miles) from Earth, or 1155 times as far as the moon and 3.13 times as far as the sun today. Radio signals, traveling at the universal limit of the speed of light, take 51 minutes to make the round trip.

Dr. Marc D. Rayman
8:00 a.m. PST January 27, 2012